Bright Light Therapy as Augmentation of Pharmacotherapy for Treatment of Depression: A Systematic Review and Meta-Analysis.

Abstract

Bright light therapy has demonstrated efficacy and is an accepted treatment for seasonal depression. It has been suggested that bright light therapy may have efficacy in nonseasonal depressions. Also, there is evidence that bright light therapy may improve responsiveness to antidepressant pharmacotherapy. We searched PubMed/MEDLINE, PsycINFO, PsycARTICLES, CINAHL, EMBASE, Scopus, and Academic OneFile for English-language literature published between January 1998 and April 2016, using the keywords bright light therapy AND major depression, bright light therapy AND depress*, bright light therapy AND bipolar depression, bright light therapy AND affective disorders, circadian rhythm AND major depression, circadian rhythm AND depress*, and circadian rhythm AND affective disorder. Studies that reported randomized trials comparing antidepressant pharmacotherapy with bright light therapy ≥ 5,000 lux for ≥ 30 minutes to antidepressant pharmacotherapy without bright light therapy for the treatment of nonseasonal depression were included. Studies of seasonal depression were excluded. Following review of the initial 112 returns, 2 of the authors independently judged each trial, applying the inclusionary and exclusionary criteria. Ten studies were selected as meeting these criteria. Subjects in these studies were pooled using standard techniques of meta-analysis. Ten studies involving 458 patients showed improvement using bright light therapy augmentation versus antidepressant pharmacotherapy alone. The effect size was similar to that of other accepted augmentation strategies, roughly 0.5. Analysis of pooled data from randomized trials provides evidence for the efficacy of use of bright light therapy ≥ 5,000 lux for periods ≥ 30 minutes when used as augmentation to standard antidepressant pharmacotherapy in the treatment of major depressive disorder and bipolar depression without a seasonal pattern.